pinctrl/amd: Configure GPIO register using BIOS settings
diff mbox

Message ID CY1PR12MB0042D650A9FCDC4B1676FB0EDCFA0@CY1PR12MB0042.namprd12.prod.outlook.com
State New
Headers show

Commit Message

Agrawal, Nitesh-kumar Sept. 9, 2016, 3:18 p.m. UTC
In the function amd_gpio_irq_set_type, use the settings provided by
the BIOS,when the LevelTrig is Edge and activeLevel is HIGH, to configure
the GPIO registers. Ignore the settings from client.

Reviewed-by: Pankaj Sen <Pankaj.Sen@amd.com>
Signed-off-by: Nitesh Kumar Agrawal <Nitesh-kumar.Agrawal@amd.com>
---
 drivers/pinctrl/pinctrl-amd.c | 14 ++++++++++++++
 1 file changed, 14 insertions(+)

Comments

Linus Walleij Sept. 13, 2016, 8:24 a.m. UTC | #1
On Fri, Sep 9, 2016 at 5:18 PM, Agrawal, Nitesh-kumar
<Nitesh-kumar.Agrawal@amd.com> wrote:

> In the function amd_gpio_irq_set_type, use the settings provided by
> the BIOS,when the LevelTrig is Edge and activeLevel is HIGH, to configure
> the GPIO registers. Ignore the settings from client.
>
> Reviewed-by: Pankaj Sen <Pankaj.Sen@amd.com>
> Signed-off-by: Nitesh Kumar Agrawal <Nitesh-kumar.Agrawal@amd.com>

Thanks! It turns out I left a partly resolved commit in the tree
so applied parts of this patch as a fixup.

Sorry for my screw-ups.

Yours,
Linus Walleij
--
To unsubscribe from this list: send the line "unsubscribe linux-gpio" in
the body of a message to majordomo@vger.kernel.org
More majordomo info at  http://vger.kernel.org/majordomo-info.html
Dmitry Torokhov Sept. 28, 2017, 6:58 a.m. UTC | #2
On Fri, Sep 09, 2016 at 03:18:09PM +0000, Agrawal, Nitesh-kumar wrote:
> In the function amd_gpio_irq_set_type, use the settings provided by
> the BIOS,when the LevelTrig is Edge and activeLevel is HIGH, to configure
> the GPIO registers. Ignore the settings from client.
> 
> Reviewed-by: Pankaj Sen <Pankaj.Sen@amd.com>
> Signed-off-by: Nitesh Kumar Agrawal <Nitesh-kumar.Agrawal@amd.com>

Just found this patch in the tree. Can you please explain why it is
needed (the patch description unfortunately tells what the patch does,
but not why).

I would expect that we either allow reprogramming the trigger as client
wishes or would error out and let the upper layers know. Silently
"fixing" the settings is wrong course of action in my opinion.

If this was trying to work around Elan touchpad drivers not working on
AMD platforms it needs to be solved in Elan driver, not here.

Thanks!

> ---
>  drivers/pinctrl/pinctrl-amd.c | 14 ++++++++++++++
>  1 file changed, 14 insertions(+)
> 
> diff --git a/drivers/pinctrl/pinctrl-amd.c b/drivers/pinctrl/pinctrl-amd.c
> index 828148d..6408dda 100644
> --- a/drivers/pinctrl/pinctrl-amd.c
> +++ b/drivers/pinctrl/pinctrl-amd.c
> @@ -384,6 +384,8 @@ static int amd_gpio_irq_set_type(struct irq_data *d, unsigned int type)
>  {
>  	int ret = 0;
>  	u32 pin_reg;
> +	bool level_trig;
> +	u32 active_level;
>  	unsigned long flags;
>  	struct gpio_chip *gc = irq_data_get_irq_chip_data(d);
>  	struct amd_gpio *gpio_dev = gpiochip_get_data(gc);
> @@ -391,6 +393,18 @@ static int amd_gpio_irq_set_type(struct irq_data *d, unsigned int type)
>  	spin_lock_irqsave(&gpio_dev->lock, flags);
>  	pin_reg = readl(gpio_dev->base + (d->hwirq)*4);
>  
> +	/*
> +	* Use the settings provided by the BIOS, when the LevelTrig is
> +	* EDGE and the activeLevel is HIGH, ignore the settings coming
> +	* from the client to configure the GPIO register.
> +	*/
> +	level_trig = !(pin_reg & (LEVEL_TRIGGER << LEVEL_TRIG_OFF));
> +	active_level = pin_reg & (ACTIVE_LEVEL_MASK << ACTIVE_LEVEL_OFF);
> +
> +	if (level_trig && ((active_level >> ACTIVE_LEVEL_OFF) == ACTIVE_HIGH)) {
> +		type = IRQ_TYPE_EDGE_FALLING;
> +	}
> +
>  	switch (type & IRQ_TYPE_SENSE_MASK) {
>  	case IRQ_TYPE_EDGE_RISING:
>  		pin_reg &= ~BIT(LEVEL_TRIG_OFF);
Agrawal, Nitesh-kumar Sept. 28, 2017, 7:46 a.m. UTC | #3
-----Original Message-----
From: Dmitry Torokhov [mailto:dmitry.torokhov@gmail.com] 
Sent: Thursday, September 28, 2017 12:28 PM
To: Agrawal, Nitesh-kumar <Nitesh-kumar.Agrawal@amd.com>
Cc: linus.walleij@linaro.org; Sen, Pankaj <Pankaj.Sen@amd.com>; S-k, Shyam-sundar <Shyam-sundar.S-k@amd.com>; linux-gpio@vger.kernel.org
Subject: Re: pinctrl/amd: Configure GPIO register using BIOS settings

On Fri, Sep 09, 2016 at 03:18:09PM +0000, Agrawal, Nitesh-kumar wrote:
> In the function amd_gpio_irq_set_type, use the settings provided by 
> the BIOS,when the LevelTrig is Edge and activeLevel is HIGH, to 
> configure the GPIO registers. Ignore the settings from client.
> 
> Reviewed-by: Pankaj Sen <Pankaj.Sen@amd.com>
> Signed-off-by: Nitesh Kumar Agrawal <Nitesh-kumar.Agrawal@amd.com>

Just found this patch in the tree. Can you please explain why it is needed (the patch description unfortunately tells what the patch does, but not why).

I would expect that we either allow reprogramming the trigger as client wishes or would error out and let the upper layers know. Silently "fixing" the settings is wrong course of action in my opinion.

If this was trying to work around Elan touchpad drivers not working on AMD platforms it needs to be solved in Elan driver, not here.

[Nitesh]We have reverted the patch in the later version of the patch.

--
To unsubscribe from this list: send the line "unsubscribe linux-gpio" in
the body of a message to majordomo@vger.kernel.org
More majordomo info at  http://vger.kernel.org/majordomo-info.html
Dmitry Torokhov Sept. 28, 2017, 5:07 p.m. UTC | #4
Hi Nitesh,

On Thu, Sep 28, 2017 at 12:46 AM, Agrawal, Nitesh-kumar
<Nitesh-kumar.Agrawal@amd.com> wrote:
> -----Original Message-----
> From: Dmitry Torokhov [mailto:dmitry.torokhov@gmail.com]
> Sent: Thursday, September 28, 2017 12:28 PM
> To: Agrawal, Nitesh-kumar <Nitesh-kumar.Agrawal@amd.com>
> Cc: linus.walleij@linaro.org; Sen, Pankaj <Pankaj.Sen@amd.com>; S-k, Shyam-sundar <Shyam-sundar.S-k@amd.com>; linux-gpio@vger.kernel.org
> Subject: Re: pinctrl/amd: Configure GPIO register using BIOS settings
>
> On Fri, Sep 09, 2016 at 03:18:09PM +0000, Agrawal, Nitesh-kumar wrote:
>> In the function amd_gpio_irq_set_type, use the settings provided by
>> the BIOS,when the LevelTrig is Edge and activeLevel is HIGH, to
>> configure the GPIO registers. Ignore the settings from client.
>>
>> Reviewed-by: Pankaj Sen <Pankaj.Sen@amd.com>
>> Signed-off-by: Nitesh Kumar Agrawal <Nitesh-kumar.Agrawal@amd.com>
>
> Just found this patch in the tree. Can you please explain why it is needed (the patch description unfortunately tells what the patch does, but not why).
>
> I would expect that we either allow reprogramming the trigger as client wishes or would error out and let the upper layers know. Silently "fixing" the settings is wrong course of action in my opinion.
>
> If this was trying to work around Elan touchpad drivers not working on AMD platforms it needs to be solved in Elan driver, not here.
>
> [Nitesh]We have reverted the patch in the later version of the patch.
>

Where was it reverted? I am looking at today's linux-next, and I am
still seeing this comment in pinctrl-amd.c:

        /* Ignore the settings coming from the client and
        * read the values from the ACPI tables
        * while setting the trigger type
        */

Thanks.
Agrawal, Nitesh-kumar Oct. 3, 2017, 4:52 a.m. UTC | #5
Hi Dmitry,

On Thu, Sep 28, 2017 at 12:46 AM, Agrawal, Nitesh-kumar <Nitesh-kumar.Agrawal@amd.com> wrote:
> -----Original Message-----

> From: Dmitry Torokhov [mailto:dmitry.torokhov@gmail.com]

> Sent: Thursday, September 28, 2017 12:28 PM

> To: Agrawal, Nitesh-kumar <Nitesh-kumar.Agrawal@amd.com>

> Cc: linus.walleij@linaro.org; Sen, Pankaj <Pankaj.Sen@amd.com>; S-k, 

> Shyam-sundar <Shyam-sundar.S-k@amd.com>; linux-gpio@vger.kernel.org

> Subject: Re: pinctrl/amd: Configure GPIO register using BIOS settings

>

> On Fri, Sep 09, 2016 at 03:18:09PM +0000, Agrawal, Nitesh-kumar wrote:

>> In the function amd_gpio_irq_set_type, use the settings provided by 

>> the BIOS,when the LevelTrig is Edge and activeLevel is HIGH, to 

>> configure the GPIO registers. Ignore the settings from client.

>>

>> Reviewed-by: Pankaj Sen <Pankaj.Sen@amd.com>

>> Signed-off-by: Nitesh Kumar Agrawal <Nitesh-kumar.Agrawal@amd.com>

>

> Just found this patch in the tree. Can you please explain why it is needed (the patch description unfortunately tells what the patch does, but not why).

>

> I would expect that we either allow reprogramming the trigger as client wishes or would error out and let the upper layers know. Silently "fixing" the settings is wrong course of action in my opinion.

>

> If this was trying to work around Elan touchpad drivers not working on AMD platforms it needs to be solved in Elan driver, not here.

>

> [Nitesh]We have reverted the patch in the later version of the patch.

>


Where was it reverted? I am looking at today's linux-next, and I am still seeing this comment in pinctrl-amd.c:

        /* Ignore the settings coming from the client and
        * read the values from the ACPI tables
        * while setting the trigger type
        */


Ok. We have reverted the code but the comments are still there. We will submit a patch to revert the comments also. Thanks for reviewing the patch.

Patch
diff mbox

diff --git a/drivers/pinctrl/pinctrl-amd.c b/drivers/pinctrl/pinctrl-amd.c
index 828148d..6408dda 100644
--- a/drivers/pinctrl/pinctrl-amd.c
+++ b/drivers/pinctrl/pinctrl-amd.c
@@ -384,6 +384,8 @@  static int amd_gpio_irq_set_type(struct irq_data *d, unsigned int type)
 {
 	int ret = 0;
 	u32 pin_reg;
+	bool level_trig;
+	u32 active_level;
 	unsigned long flags;
 	struct gpio_chip *gc = irq_data_get_irq_chip_data(d);
 	struct amd_gpio *gpio_dev = gpiochip_get_data(gc);
@@ -391,6 +393,18 @@  static int amd_gpio_irq_set_type(struct irq_data *d, unsigned int type)
 	spin_lock_irqsave(&gpio_dev->lock, flags);
 	pin_reg = readl(gpio_dev->base + (d->hwirq)*4);
 
+	/*
+	* Use the settings provided by the BIOS, when the LevelTrig is
+	* EDGE and the activeLevel is HIGH, ignore the settings coming
+	* from the client to configure the GPIO register.
+	*/
+	level_trig = !(pin_reg & (LEVEL_TRIGGER << LEVEL_TRIG_OFF));
+	active_level = pin_reg & (ACTIVE_LEVEL_MASK << ACTIVE_LEVEL_OFF);
+
+	if (level_trig && ((active_level >> ACTIVE_LEVEL_OFF) == ACTIVE_HIGH)) {
+		type = IRQ_TYPE_EDGE_FALLING;
+	}
+
 	switch (type & IRQ_TYPE_SENSE_MASK) {
 	case IRQ_TYPE_EDGE_RISING:
 		pin_reg &= ~BIT(LEVEL_TRIG_OFF);